MacLeod and Hodder (1998) demonstrated that presenting two different incongruent color words in the same color on a single Stroop trial resulted in no more interference than did presenting the same incongruent color word twice, and concluded that the first word captured attention, blocking out the second. They also showed that, within a trial, neither stimulus onset asynchrony between the two items nor the presence/absence of a visible gap between the two items had any effect. We replicated all of their empirical findings. Then, by extending their design and factorially combining three types of items--incongruent words, congruent words, and control nonwords--within a trial, we demonstrated that both items within a trial do influence processing, with the contribution of the second greater than that of the first. These results are incompatible with a capture account and suggest instead that the word dimension continues to be monitored during the attempt to identify and produce the name of the color.